A Dimension-Independent Generalization Bound for Kernel Supervised Principal Component Analysis.
Hassan AshtianiAli GhodsiPublished in: FE@NIPS (2015)
Keyphrases
- principal component analysis
- feature space
- kernel pca
- principal components
- upper bound
- independent component analysis
- lower bound
- learning algorithm
- kernel principal component analysis
- face recognition
- supervised learning
- dimensionality reduction
- linear discriminant analysis
- image processing
- real valued functions
- kernel methods
- semi supervised
- support vector
- singular value decomposition
- covariance matrix
- unsupervised learning
- dimension reduction
- multiple kernel learning
- kernel function
- gaussian processes
- supervised classification
- random projections
- kernel matrix
- reproducing kernel hilbert space
- objective function
- constructive induction
- feature extraction